ABSTRACT Alzheimer's Disease (AD) represents a devastating disease characterized by the presence of protein aggregates within the brain that are closely linked to synaptic nerve loss and progressive cognitive deficits. A lifestyle intervention, such as exercise, can improve cognitive function in AD patients. While the mechanisms that mediate this change are likely multifactorial, our proposal investigates the potential for a novel cell-to-cell signaling mechanism via transfer of extracellular vesicles (EV; exosomes and microvesicles) to mediate changes in the neuronal cells. Our preliminary data show that aerobic exercise can both increase the number of EVs in serum as well as their content of heat shock proteins (HSP). Moreover, we demonstrate that EVs with elevated HSP can inhibit protein aggregation within neuronal cells. We provide additional evidence suggesting this EV clearance uses an autophagosomal pathway. Our central hypothesis is exercise induced EVs can impede the protein aggregation typically associated with AD. In this proposal we will test this in two independent specific aims. Aim 1 will assess whether exercise of AD patients modulates EV content, numbers and function (i.e., ability to block protein aggregation). Aim 2 will assess the molecular underpinning through which exercise- induced EVs and their associated HSPs might be working. In this latter aim, we will test whether the autophagosomal pathway is involved in the removal of amyloid-?-peptide (A?) containing protein aggregates. This proposal represents an innovative approach to study AD, and has the potential to provide insight into how exercise can improve AD outcomes and ultimately may provide novel therapeutic approaches.